> companies are generally cautious about this sort of thing
yeah, they hire lobbyists to do it for them so they don't have culpability, but you are right.
363 karma · joined December 2, 2022
yeah, they hire lobbyists to do it for them so they don't have culpability, but you are right.
the real world standard is don't do anything that knowingly or intentionally violates the law or your employers policies.
anyone can falsely claim that what you are doing is wrong but as long as it doesn't break employer policy or law then they are wrong.
in the corporate world it would be suicidal to "avoid the appearance of impropriety" because your competitors will slaughter you while they are following the letter of the law.
500 consecutive hours should be far superior because the time is focused and with fewer distractions.
training people in a professional 9-5 corporate environment as opposed to the adult daycare resort environment of universities would probably lead to better outcomes as well.
very little of what you learn in comp sci is relevant to your job as a new grad.
for software engineering a comp sci new grad still needs years of mentorship, supervision and on the job training in order to perform at a senior level in whatever specialty they end up working in.
because software engineering is so highly specialized pretty much everything you need to know for a particular job must be learned on the job.
the only thing a comp sci degree really shows is that a person has interest in the field and is capable of learning, but there are much better ways to figure that out than requiring degrees.
humans have been doing it for 1000's of years. that is a much longer record of sustainable success than anything else we know of.
there isn't any particular reason to think that industrial agriculture will fail any time soon. the worst case scenario is that it becomes more expensive over time. we spend <5% of our GDP on food, so even a 10X increase in raw production cost would be survivable.
we massively over produce food (30% of corn burned as gasoline off the top, another large percentage of soy and corn used as animal feed, huge amounts of waste) and the biggest nutrition problem we have is obesity.
the US has huge amounts of underutilized marginal land simply because there is no reason to use it. we already have way too much food.
not as long as they can force people to pay for it themselves and have the government and the tax payers subsidize it.
universities were never meant to be vocational schools though, and they do a horrible job at it.
to take computer science as an example: a 4 years bachelor degree only has about 500 hours of classroom instruction in comp sci subjects.
a 9-5 vocational training program can provide 480 hours of classroom instruction in 3 months.
the idea that students benefit by going to a university where they only spend a few hours a day learning and the rest of the time playing games, socializing, drinking and doing drugs, is absurd.
put people in a 9-5 40 hour a week program that has the same attendance and performance requirements as the job they are training for and they can be trained to the level of any bachelors degree in 3-6 months.
so why not just create a test that actually tests for the skills needed for the job? that would be the obvious way to optimize profit from a business perspective.
the fact that employers don't do that is a pretty clear sign that the purpose of degree requirements is explicitly discriminatory, the same as "IQ" tests (which are completely fraudulent).
that is the only way to eliminate the systemic discrimination that the university system and degree requirements propagate.
almost all serious jobs already have objective skills tests and licensing requirements for employment: doctors, lawyers, financial services, heavy equipment operators, etc, etc, etc.
adding degree requirements on top of those licensing requirements cannot possibly add any additional benefit but it is obviously discriminatory against many marginalized groups.
any source on that? as far as i know there was never any official roadmap or announcement by tsmc of gaafet on n3.
this article seems to be from the n3 announcement (april edit: 2020) (https://www.techpowerup.com/265991/tsmc-3nm-process-packs-25...) and it was announced as finfet then.
another article from n3 announcement in april 2020: https://seekingalpha.com/article/4338843-taiwan-semiconducto... ("Taiwan Semiconductor Manufacturing Company Losing Its Process Leadership To Intel")
intel has consistently failed to deliver on their prior roadmaps and is currently behind tsmc in what they are shipping, so it requires a bit of magical thinking to believe that they will suddenly get a 2 year lead on tsmc.
intel is still not shipping 4, which is their first part to use euv. based on what they are shipping, they cant do it at volume yet. the idea that they are going to scale euv and then jump to gaafet and bspd in the next 18 months is kind of beyond belief.
when intel was failing to deliver 10nm (now called 7) in volume, they were saying the exact same thing, that 10nm was delayed but 7nm (now called 4) was on schedule (https://www.anandtech.com/show/15217/intels-manufacturing-ro...). back then intel 4 was scheduled for 2021. the magic leap didnt happen and it is 2 years late so really no reason to think that everything else wont get pushed back the same this time.
tsmc has already been shipping euv products in volume for years and they have a roadmap for introducing gaafet and bspd incrementally which looks a lot more realistic.
However, fabs get judged on what they ship, not what the promise. TSMC is shipping N3 in volume, and Intel 4 is still not shipping, and sounds like it will be a cut down limited volume release (https://www.tomshardware.com/news/intel-meteor-lake-cpus-for...) most likely because the yields are not there.
Based on prior delivery it is highly unlikely that Intel will ship meaningful volume on 20A before TSMC does on N2, but miracles happen.
the bigger sample bias is that wealthy people almost universally send their children to college and their children are always going to earn above average wages.
it is pretty obvious that people who graduate college are going to earn more because large numbers of high paying jobs require a college degree, but that is a bad thing.
jobs that require specific skills should have objective skills tests to qualify for the job. requiring a degree for a job should be prohibited by law because it is blatantly discriminatory.
AI today is built using classical computers but the human mind is a quantum computer. AI is a simulation of a quantum computing device built using a classical computing device. The simulation might provide some insight into how the real thing works but the underlying mechanism is completely different.
if ChatGPT is good at cheating all that proves is that the curriculum and grading mechanism are garbage, which is already a well known fact, so shouldn't come as any surprise.
a school based on dialog, exploration, experimentation (e.g. true learning) could never be "cheated" with a toy that regurgitates thing it read on the internet based on probabilities.
just wait until they are abandoned by bankrupt companies and start to break down and all of the highly toxic substances they are made of leech into the ground water.
as mentioned, everything fossil fuel related has a massively higher death and injury cost than nuclear.
the california wildfires caused by the criminally negligent failure to maintain power distribution systems killed more people and cost more money that Chernobyl (look it up).
The paper you link to is a joke.
They claim a cost of utility scale solar as $17 / MWh and then claim a cost of utility scale solar with battery as $22 / MWh.
How much battery backup does that $5 / MWh provide? it doesn't say, but the answer is only a few minutes.
Look at the chart of page 28 of planned installations where it shows 22 GW standalone storage being added for 60 GW of solar and wind. That means the standalone storage being installed can provide 20 minutes of backup power for the renewable resources being installed.
The cost of a renewable system with battery backup that can guarantee power for even 24 hours would probably be at least 10X the cost of bare solar, so in the neighborhood of $200 / MWh, or quadruple the target cost of NuScale (based on the chart in this paper).
Guaranteeing power for 24 hours is not acceptable though. A system like that would have regular blackouts given the normal variability in solar and wind.
Unless someone is quoting a price on guaranteed sustained power delivery they are not serious. Try selling your electricity to a data center or any other commercial buyer and see how far you get.
i think this guy understands the industry very well. maybe you are the one with tik-tok opinions?
so, anywhere where your workloads or data are physically co-located on the same hardware as someone else's should be automatically disqualified, right?
so, if mercury has a 1M FDIC guaranteed, then you can move 1M per day through their systems with 100% guarantee, but the risk that a bank fails on one particular day is very small, so maybe you can move 10M a day through them, with effectively no risk.
10M a day is 3.5B a year. very few companies are doing transactions on that level, so mercury scales up very well.
the underlying issue was that SVB knew they were f'd and so they were begging VCs to stop taking money out because they knew they couldnt cover and those VCs turned around and pulled money faster, because that is what smart money does.
SVBs bargaining chip was that the VCs wouldnt want the books being opened, which is what happens when the fed takes over, but the VCs called their bluff, because they figured they were well enough insulated from any potential fall out.
this is america, so we have the keating S&L scandal, ltcm, enron, worldcom, countrywide financial, 2008. the game never stops.